castanyes blaves

Random ramblings about some random stuff, and things; but more stuff than things -- all in a mesmerizing and kaleidoscopic soapbox-like flow of words.


verbatim latex






The biobar project is a one-stop toolbar that provides access to all major biological data resources. The primary advantage of this tool is that it allows a biologist to browse and retrieve data from Genomic, Proteomic, Functional, Literature, Taxonomic and Structural databases. The tool also provides links to major data deposition sites for nucleotide, protein and 3D-structure data and contains links to many Sequence, Structure alignment and analysis tools. Databases supported include NCBI, DDBJ, and EBI databases.


The biobar project is a one-stop toolbar that provides access to all major biological data resources. The primary advantage of this tool is that it allows a biologist to browse and retrieve data from Genomic, Proteomic, Functional, Literature, Taxonomic and Structural databases. The tool also provides links to major data deposition sites for nucleotide, protein and 3D-structure data and contains links to many Sequence, Structure alignment and analysis tools. Databases supported include NCBI, DDBJ, and EBI databases.

Wind blowers

We went for a mountain bike ride to Rubió. There is a new Aeolic Park there:

With quite mastodontic turbines:

There are concerns from the environmentalist point of view because most of the birds are unable to see the helixes, and they got killed. But it is much better than most of the other energy sources used in Catalunya, with coal, gas and fuel being the worst.


Something to do tomorrow

Get Luminocity running in my Hoary Laptop

IronPython has moved to a "Shared Source" license

Jim Hugunin released version 0.7 of IronPython and moved to a "Shared Source" license. Some people is complaining about this and already leaving the project and looking to other (play)grounds, namely Boo.

There has already been a discussion about the decision to license IronPython as "Shared Source" from now on. Jim Hugunin doesn't see any problem in that:

He points to Mr. Matusow, at Microsoft, who is the one in charge of the Shared Source stuff, for the license discussions:

What will happen if Microsoft, Jim Hugunin's employer, locks in IronPython?

Will IronPython succeed as a dynamic programming language for the Common Language Infrastructure?

Will Boo, already having static typing, replace IronPython?

Time will tell...


Every Monday, Si sets you a riddle to get your brain working.

Another Age Old Problem

Last week was a lady's birthday. At her party that evening she set this challenge: "I have three children. The sum of their ages is equal to my age and the product of their ages is equal to the year I was born in. How old are they?"

Some more Mono / Beagle / Xlg+Luminocity goodies


programming c-sharp 4th ed notes

Wow! This is what I call a good start

Programming C#, Fourth Edition, will work with Mono

(page xii, Preface)

ubuntu faster bootups when laptop has no networks near

Delete the "auto eth0", "auto ra0", etc, entries in /etc/network/interfaces

Then do:

sudo mv /etc/rcS.d/S40networking /etc/rcS.d/NOT_S40networking

sudo mv /etc/rcS.d/S51ntpdate /etc/rcS.d/NOT_S51ntpdate

You will have to manually "ifup" the interfaces once you arrive at a place with net:

Connect the device (ethernet, wireless card, ...)

/sbin/ifup {eth0, ra0, ...}



All these commands require sudo!

* apt-get update - Run this after changing /etc/apt/sources.list or /etc/apt/preferences. You also must run it periodically to make sure your source list is up-to-date.

* apt-get install packagename - installs a new package (but see aptitude, below)

* apt-cache search string - Searches for string in the list of known packages

* dpkg -l package-name-pattern - List packages matching pattern

* aptitude - Curses viewer of packages installed or available. Aptitude can be used from the command-line in a similar way to apt-get, but only for some commands - install and remove being the most common. However, because aptitude keeps track of more information than apt-get does, it can be considered better at install and remove operations.

* apt-cache showpkg pkgs - Show information about packages.

* apt-cache dumpavail - Prints out an available list.

* apt-cache show pkgs - Displays package records, similar to dpkg --print-avail.

* apt-cache pkgnames - Fast listing of every package in the system.

* dpkg -S file - Which installed package owns the file?

* dpkg -L package - List files in the package.

* apt-file search filename - Search for a package (need not be installed) containing files including the string. apt-file is a package of its own, which you may have to apt-get install first, then run apt-file update. If apt-file search filename shows you too much, try apt-file search filename | grep -w filename (which shows you only the files that contain filename as a whole word) or variants like apt-file search filename | grep /bin/ only files located in directories like /bin or /usr/bin, useful if you're looking for a particular executable).

* apt-get autoclean - Run this periodically to clean out .deb archives from packages which are no longer installed on the system. You can regain lots of disk space that way. If you're really desperate for disk space, apt-get clean is more radical, and will remove .deb files even for packages currently installed. But most of the time you probably don't need the .debs any more, so it might be worth it if you're strapped for megabytes.

rt2500 wireless in hoary

Compile and install the GPL drivers:


make install

Make sure to

/sbin/depmod -a

Add /etc/modprobe.d/rt2500

alias ra0 rt2500

Then in /etc/network/interfaces

# rt2500

iface ra0 inet dhcp

name Rt2500 Wireless card

wireless_channel 11

wireless_mode managed

wireless-essid default

That should do it.

sudo /sbin/ifup ra0

Check that you have it in System->Administration->Networking


Emacs dired mode

Create a directory:


Immediate -> Create directory


Beagle 0.0.8

Beagle 0.0.8 represents a huge step forward in stability, and contains fixes for a multitude of bugs and memory leaks. This is the first version of Beagle that is really suitable for everyday use. It also includes a brand new web services API to the Beagle daemon.

To download the 0.0.8 tarball, visit the Beagle web page at

Programming C# 4th Ed

Today I have started reading this masterpiece of C# 2.0 and .NET 2.0 book by O'Reilly, that comes right in time for Easter holidays.

Just one day after my birthday: thank you O'Reilly!


in girubimus nocte et consumibur igni

Sobre las putadillas de la beta, es que cuando arrancas el GDM se vuelve loco y has de hacer todo el proceso en módo cónsola. Luego, importar a mano las llaves GPG que hay en el CD y luego cascarle al update de turno. Pero por lo demás, la cosa va tirando. Sigo sin soporte MP3 o de vídeo decente, pero ya sabemos el pollo que hay con las patentes y las licencias.

Y por último, se ve que se pueden usar los repositorios de extras en esta beta, por lo que se podrán instalar algunos programas que por falta de espacio no han entrado en la versión final de Fedora. Para instalar esto, simplemente hay que añadir estos ficheros en el directorio /etc/yum.repo.d/

name=Fedora Extras - development - $basearch




Today I jump into the 27 club

Anybody reading this has a beer/any-other-beverage paid. Send an email to my gmail account or give me a call (to the usual numbers).


PC Peru

Gobierno peruano presenta «PC Peru» con sistema GNU/Linux

Sección: Gobiernos

18/Mar/2005 - 02:21

El gobierno peruano a través del Ministerio de la Producción ha formado una alianza con la Corporación Intel para que la población pueda adquirir computadoras personales a precios accesibles, con el fin de promocionar el uso de tecnologías de la información en los sectores productivos y de servicios.

Esta máquina económica viene con el sistema operativo GNU/Linux y software libre preinstalados. Su precio es de $400 en la versión económica.

Expressions I should be using more often

Nothing could be further from the truth

Capellades computer shops

(1) MicroInfo

(2) a second one nobody remembers its name




I went to visit the new Cosmocaixa this afternoon. To sum it up, new building, mostly the same sets.

But, what I really enjoyed, and wasn't expecting a lot, is the new selvatic (Amazonian?) forest reproduction (see photo): with water falling from the ceiling, simulating rain, they have a nice place that you can look through almost everywhere.

Plus, they have this great warning message by the pool that says: "don't put your hand inside the water, some of the fishes are carnivorous and will look at you hand as if it were a delicious family of five worms...". This being in a country unlike Sweden or Holland, they have a guard remembering people not to do exactly the other way around what it is explained in the message... beware that some of the fishes where almost a meter long, and that there is an (sleepy) anaconda of about 3m long.

They also have a section for kids, called "toca toca" (touch touch!). This is great for children wanting to interact (translate: try to break) with the scientific sets.

Yet, in the normal (adults+kids) museum, you have to be alert of the "kamikaze look-ma kids": while you are absorted with a set, staring at it and pushing the buttons, this kami will run right into you, push you out, cut your hand, or stab an elbow at your belly, while shouting: "look ma!!! run over here!!!". Again, if this where the British Museum in London, and a kid would behave like that, they would be dealing with it the next morning in the upper chamber House of Lords...



command up2date, or up2date --nox for systems not running the X Window System.

seq command

[~]$ seq 1 10 | perl -p -e '$_ = "prefix: $_"'

prefix: 1

prefix: 2

prefix: 3

prefix: 4

prefix: 5

prefix: 6

prefix: 7

prefix: 8

prefix: 9

prefix: 10

Yahoo! Buzz! Game!

I! have! started! playing! Yahoo! Buzz! Game!

Buying! $5000! shares! of! Ubuntu! and! $5000! shares! of! Fedora!

Let's! see! what! future! departs!


Intel globalises Linux programme,39020390,39191513,00.htm

The chipmaker is to give 160,000 of its sales channel partners access to software kits that make pre-installation of the open source OS easier

Intel, which launched a programme in November to make it easier for Chinese and Indian computer makers to sell Linux on PCs, has now made the program global, the company said on Tuesday. That means that about 160,000 sales channel partners will get access to Intel's software kits for Linux, which include supporting software, scripts to automate installation and compatibility-checking tools, Intel said.

"The demand for Linux on the desktop, while still small, is growing and represents an opportunity for Intel's channel network," Intel spokesman Scott McLaughlin said in a statement. The software kit works with Linux from Red Hat, Novell, Red Flag and China Standard Software. Intel's corresponding programme for Microsoft Windows is much better developed, but Intel said it will improve the Linux kit.

Linux drives Renault Formula 1,39020381,39191523,00.htm

The Renault Formula 1 team is running simulations and crucial telemetry applications on Linux clusters from IBM, and it's pleased with the results.

The IT team behind the Renault Formula 1 team has dramatically cut the time it takes to test new features by using Linux, Renault said on Tuesday.

Renault F1 is running Linux on an IBM e1350 cluster, based on IBM eServer e325 and e326 servers with AMD Opteron processors.


Expression I should be using more often

This highlights a hitherto unforeseen contribution of genomic location to coding sequence evolutionary constraint

print non-comment lines

grep -v "^#.*"


referencing equations in latex


\hat{\mu}_{k,r} = \max_{i \leq k} \, \min_{j \geq k} \frac{X_{i,r} +

\ldots + X_{j,r}}{j - i + 1}.



and then use:

The estimated trend

(Equation~\ref{eqn:isotonic_trend}) is nondecreasing...

German Railways Achieve Major Milestone in Strategic Move to Linux

IBM press release

With the Successful Move of 55,000 Lotus Notes Users to Linux, the First Milestone in the Deutsche Bahn's Strategic Enterprise-Wide Linux Migration Has Been Achieved

FRANKFURT, GERMANY -- Feb 2, 2005 -- Deutsche Bahn, the company that manages the German railway system, has successfully completed the migration of their Lotus Notes system to an IBM eServer zSeries 990 running Linux.


and add this line in your config file

In your savage driver section in /etc/X11/xorg.conf

Option "DmaMode" "Any"


Install the new savage-20050311 snapshot

Felix Kuhling and Alex Deucher have headed the cooking of a gracious new release of the savage dri drivers.

If you happen to have a laptop or desktop with a savage graphics card, and you enjoy an occasional game with The Open Racing Car System or other FPS-churning game, you have to try it:

$ wget

$ su

# /sbin/init 3

# tar xjf savage-20050311-linux.i386.tar.bz2

# cd dripkg

# ./

(Return - Return - Return - Return)

# /sbin/init 5

And voilá...


Those who have a smile in their face are those who haven't listened the latest news yet

After a year of a horrible point in history:

reuters - Spain solemnly commemorated the first anniversary of the Madrid train bombings on Friday with church bells and silent tributes to the 191 people who died in al Qaeda's worst attack in Europe.

bbc - The government initially blamed Basque separatists Eta for the attacks, an error which is widely believed to have contributed to the Popular Party's unexpected defeat.

The PP is still convinced there could be a further link to Eta - and this has led to rancorous disagreement with the governing Socialists.

Responsibility for the attack was claimed by a Moroccan cell with links to al-Qaeda, and most of those who have been arrested are Moroccan citizens.

Reducing memory usage in freetype

David Turner (FreeType Project) has been working on reducing the memory usage of freetype, specially when truetype fonts are used.


An error in the system

7 March 2005 -- This morning, the Council Presidency rubberstamped the illegal patenting practise of the European Patent Office by adopting a proposal without vote and without a qualified majority of member states, "so as not to create a precedent which might have a consequence of creating future delays in other processes". Jonas Maebe, Belgian computer scientist and board member of FFII, explains in an open letter how the experience with the software patent directive has proven the EU and its proposed Constitution to be a dangerous political farce.

Dear Legislators of the European Union,

Dear Supporters of the European Constitution,

I'm not "against Europe", nor against the concept of the EU. On the contrary, I very much enjoy several of the benefits that the EU has brought us and I think that close cooperation between member states can help everyone. I'm also not against the principle of a European Constitution. That said, I would like to say a few things about democracy in Europe as I experienced it until now, the proposed Constitution and how the latter may or may not affect the former.

I have now been involved for almost two years in the discussion and political process surrounding the software patents debate. I was not involved from the start, but I did spend several weeks in the European Parliament and have been in contact with several parliamentarians, civil servants and representatives of the European Patent Office. I even spent a night in the European Parliament in Strasbourg, helping to prepare the FFII voting list for the legendary 24 September 2003 EP vote. Overall, it has been a very educational two years.

As you are undoubtedly aware, the EU Council of Ministers adopted its Common Position on the software patents directive on 7 March 2005. You can find the Foundation for a Free Information Infrastructure's account of what happened here. There is also a quote from me on that page, in which I deride supporters of the Constitution, without providing any arguments as to why I attack the Constitution in particular.

In what follows, I would like to clarify my position.

The Council session of 7 March 2005

An MEP who saw our press release noted that "it certainly appears that the Council presidency took some procedural shortcuts". That is an extremely diplomatic way to put it. The Council Presidency spoke in name of the entire Council (pluralis majestatis as it were) and reversed voting requirements. The rules states that "an A item (formal point) shall be taken off the agenda if a member state so requests or if it might lead to further discussions, unless the Council decides otherwise".

When Denmark asked for that, the Presidency simply announced it did not want to take it off, instead of asking whether there was a majority against taking it off. Therefore, "unless the Council decides otherwise" became "if a majority of the Council wants to go against the Presidency's will, while not even a vote has been called". This might seem like splitting hairs, but in the Council where every sign of opposition is presented to us as something which foreshadows the end of EU decision making as we know it, this is a difference between night and day.

Some more interesting points:

* The request for a B item (discussion point) was supported by Poland, Portugal, the Netherlands and Denmark. Losing the support of even one of the first three of those countries was enough to be left with a text no longer supported by a qualified majority. Asking for renewed discussions means that they are not happy with the current text, otherwise they wouldn't do so.

* Unilateral declarations with concerns of no less than 8 countries (7 of which officially voted in favour of the text) are attached to the text of the Common Position. Especially the Polish one is very harsh, and basically states "we recognise a political agreement has been reached and do not dare to attack it, but the text from 18 May is our worst nightmare". You can read it at

This has nothing to do with software patents. There was simply no qualified majority (possibly not even a simple majority) in the Council for this text. It was purely due to diplomatic inertia and fear of doing something against whatever is customary that it slipped through. Unless the Constitution says somewhere "the written rules always have precedence over diplomatic customs and fears", it won't change this.

National Parliaments

An often touted advantage of the Constitution is that national parliaments must get the opportunity to look at all legislation before Council deliberations. In fact, this is even used as justification for giving the Council the great power that it has. The problem with this is that this directive has shown that various governments simply do not care about what the national parliaments say (let's not even mention the European Parliament).

Before the political agreement of May 2004 was reached

The Dutch national parliament was misinformed about the directive. Minister Brinkhorst told the Dutch Parliament in April 2004 that there was a compromise between the Council and the European Parliament, so that the political agreement in the Council was a formal non-issue. One could argue that the Dutch Parliament should have verified whether or not this is true, but as Mr Rocard would put it, that was quite an "inelegancy".

Consequently, in July 2004, the Dutch Parliament adopted a motion in which it stated that it had been misinformed, and called upon the government to from then on withhold its support for the Council text. The Dutch government promised to execute this motion, but reinterpreted it as meaning "only if it ever becomes a B item again and it is voted upon, then we will change our vote into an abstention". Given all the panic reactions we saw the last few months about the possibility of this ever becoming a B item again, this amounted to saying "Nice motion, but no cigar".

At the May 2004 Council session

As you undoubtedly know, a political agreement has not an ounce of legal value in any way. As the name implies, it's purely political. Given that in this case there were three last-minute amendments introduced by the Commission and one last minute "compromise" by the German delegation and the Commission, it is fair to say that not a single parliament has had the chance to decide about the final text that was adopted on 18 May 2004. And due to everything that happened later, they didn't have any chance later on either.

Before those amendment were introduced, the majority of the Council members were against the text. After those "compromises" (which did not change anything to the substance of the text), a 10 minute break was held in which everyone was scrambling to call national experts to judge those texts. Several delegations failed to reach anyone and as such did not really know what to do. Many simply followed Germany, since they had been sort of coordinating the opposition in the Council before.

However, Germany seemed to be in bed with the Commission now, being content with the sole addition of the word "new" in the definition of "technical contribution". After Denmark was persuaded to be "80% happy" by the Council Presidency, the Presidency even didn't ask Poland anymore because their vote was no longer needed. If you have not yet seen the Denmark-Ireland dialogue, you can find it at the link below this paragraph. It's only 45 seconds long, and if it weren't about a decision as important as this one it would even be quite funny:

[ Windows media version | Quicktime/mpeg4 version ]

The European Parliament

Another advantage of the proposed Constitution is that the European Parliament has to approve the legislation in all areas in which the Council acts by majority. Since this directive is being handled under codecision already, that will not prevent any situations as they occurred in the case of this directive however.

It indeed seems like the only way out now is a massive rejection by the European Parliament in second reading, but I would not consider that a victory for the EP. It's giving up against the Commission and Council who seem to be determined to do whatever they like, unless the EP simply stops the whole procedure by destroying the directive project.

The EP can merely act as an emergency brake in the current situation, and that will not change with the Constitution as far as I can see. Their first reading will remain merely "advice" to the Council, and the second reading will still be handicapped by "majority of its component members" requirements. Additionally, the Commission can still basically nullify the EP position in the Council by disagreeing with amendments, and thus require the Council to act by unanimity rather than by qualified majority on those points.

It must be noted however that if a directive goes through until conciliation, then the EP may be able to assert itself after all. If approved a similar set of strong amendments twice before (in first and second reading), that may give them a strong negotiation position against the Council and Commission. The absolute majority requirement makes this different to achieve, but on the other hand strengthens the signal if it actually is achieved.

The Constitution

Relating to the Constitution in particular, I would like to note my concern regarding one specific paragraph, which simply says "Intellectual property shall be protected" (Article II-17). Given that many people consider software patents to be intellectual property, this almost seems to make any directive excluding software patents to become unConstitutional.

The term "intellectual property" should at least be defined in some way, because everything but the kitchen sink is categorised under that generic term (patents, copyright, trademarks, design rights, digital rights/restrictions management technologies, ...), and things keep getting added. The fallacy that "every idea" has to be someone's "intellectual property" is promoted more and more, which means that such a generic provision is extremely dangerous and may start to conflict quite severely with Article 10 of the European Charter of Human Rights in the near future.

I understand I'm quite late with my remark, but as explained before I was busy with other important things as well.


I am sorry to be so pessimistic and to throw this out all over you, even though many of you personally have no fault in all this. I'm also certain you have the best intentions with the proposed Constitution. However, I'm becoming tired:

1. We got almost unanimous support in the European Parliament's Legal Affairs Committee to restart the directive.

2. This request was confirmed by unanimity in the Conference of Presidents and an overwhelming majority in plenary

3. The Commission declined and until today has not been able to produce any explanation as to why (except for "we want the procedure to continue")

1. We managed to secure a generous blocking minority in the Council (Spain, Austria, Belgium, Italy, Portugal, Denmark, Poland, Hungary, Latvia, The Netherlands)

2. This position was confirmed by the German, Dutch, Spanish and Danish national parliaments

3. The Council presidency "takes some shortcuts" and shoves it through as a "non-discussion" item, "so as not to create a precedent which might have a consequence of creating future delays in other processes"

Because democratic ways fail over and over again, the situation has now become so bad that some people even set up a web page where you can pledge money to bribe the Council, because that's the only way they see that's left to get anything done at all. It may seem like a joke, but after everything I've experienced the past one-and-a-half year (since the directive was passed from Parliament to Council), it would not surprise me in the least if they're half-serious.

The Constitution merely enshrines all of the above. Of course it does not codify the fact that the people who lead the European Patent Office should be the same people as those who write the Council version of the directive as it happened last year, but it also does not prevent this. It does not say that the Commission should introduce last minute amendments at Council sessions so as to confuse delegations, but it also does not call a halt to this practice. It does not say political agreements are cast in stone, but neither does it clearly say that they have less legal value than the ticket you get in a supermarket and that they should be treated as such when there is reason to do so.

How on Earth am I expected to still believe in this farce? I really do want to believe. Just give me chance to do so...

Sincerely yours,

Jonas Maebe

jmaebe at

FFII Board Member

Research Assistant in Computer Science, Ghent University, Belgium

Expressions I should be using more often would be expected were this process to occur via [retrotransposition].


LinuxBIOS, is a modified version of the kernel, Linux. However, few computers can run them. Whereas "PC clones" were and are quite similar, and fully-documented as regards what the kernel and user-space programs need to know, the commands that the BIOS must execute in order to initialize the machine are varied, and in most cases secret. How to install a new BIOS is also secret on many machines. And so far, most manufacturers have not given the FSF the necessary cooperation of providing these specifications. Some desktop machines can run a free BIOS, but we don't know of any laptop that can do so.

The FSF uses laptops donated by IBM over the past few years. This was one among several ways IBM cooperated with the GNU Project. But the cooperation is incomplete: when I asked for the specifications necessary to make LinuxBIOS run on these laptops, IBM refused—citing, as the reason, the enforcement of "trusted computing" Treacherous computing is, itself, an attack on our freedom; it is also, it seems, a motivation to obstruct our freedom in other ways.

Not all of our community perceives the non-free BIOS as an acute problem. Much of our community supports the open source philosophy, which says that the issue at stake is choosing a development model that produces powerful, reliable software. The open source philosophy doesn't say that "closed source" software is unethical, only that it is likely not to be as reliable. People who hold those views might care about the loss of freedom imposed by a non-free BIOS, because in their philosophy, freedom is not the issue. For us in the free software movement, freedom is the main issue; we have to solve this problem, whether they help or not.

How You Can Help

Since requests for manufacturers' cooperation have not solved the problem, another approach is needed. Now we are asking you to help.

For instance, simply installing a new BIOS in the machine is a substantial challenge. Most manufacturers don't publish the information on how to do this. If you can figure this out for some recent model, especially a laptop, that would be a substantial contribution.

Cooperation from the manufacturers would make that work unnecessary. However, to gain cooperation we need to press for it.

The most uncooperative company is Intel, which has started a sham "open source" BIOS project. The software consists of all the unimportant parts of of a BIOS, without the hard parts. It won't run, and doesn't bring us any closer to a BIOS that does run. It is just a distraction. By contrast, AMD cooperates pretty well.


The Consensus CoDing Sequence (CCDS) project is a collaborative effort to identify a core set of human protein-coding regions that are consistently annotated and of high quality. The long-term goal is to support convergence toward a standard set of gene annotations on the human genome.

Annotation of genes on the human genome is provided by multiple public resources using different methods, resulting in information that is similar but not always identical. The human genome sequence is now sufficiently stable to start identifying gene placements that are identical, and making this data public and supported as a core set by the three major public human genome browsers. The long-term goal is to support convergence towards a standard set of gene annotations on the human genome.

Toward this end, the Consensus CDS (CCDS) project was established. The CCDS project is a collaborative effort to identify a core set of human protein-coding regions that are consistently annotated and of high quality.

The initial CCDS data set, containing nearly 15,000 genes, has been posted

on three Internet sites:

- the UCSC Genome Browser (

- the Ensembl Genome Browser (

- the NCBI CCDS Database website (


artichoke crops lost

There has been a major lost of the artichoke crops due to the several freezing nights we have recently had.

This is very bad news, as I'm an artichoke fanatic...


There is something about Abiword

After an intense debugging session by Martin Sevior, he got tight wrapped objects for AbiWord working well enough to commit.

Screenshot with a tightly wrapped object.


Expressions I should be using more often

"I'm never going to make it to the cordon bleau"

"A little rough around the edges"

Funny: "I don't want to appear selfish but, stop what you are doing

and focus on me"


Things like this make my griiiiiin

Kudos to again!

Ethereal: network protocol analyser

If you happen to be interested in what I write in this blog, and if you happen to live near Igualada (Barceona-SPAIN), check this out:

Ethereal: network protocol analyser



12h Dissabte 5 Març 2005


Sala Metropolis - La Kaserna. Trav. Sant Jaume s/n (antiga policia



Aleix Solé Romeu - iglu

Expressions I should be using more often

"When you get a moment I would like to have a word with you"

"I won't be holding my breath on that"


expressions I should be using more often

In a well-thought-out study that was designed to finally settle the arguments


Hero of the month: Ben Maurer

I have been looking into the question `where does gnome-settings-daemon use all of that memory'. After putting valtrind on it (which was no easy task, given how bonoboized it is), I found that gstreamer was allocating a whopping 300 kb at desktop startup.

What it seemed to be doing was loading some large xml (bloat sign #1: large xml document) on initialization (bloat sign #2: said large documents loaded at startup), copying it to memory (bloat sign #3: said large documents being copied with malloc).

Federico encourged me to find out exactly why this was happening. What I found simply shocked me.

We were using this in a file called `acme-volume-gstreamer.c'. Yes, we are loading 300 kb of data from an xml file at bootup time to control the volume of the computer.

In other news, gstreamer and gnome-settings-daemon are not the only bloatware on the block. Some of the glutinous usages of memory found over the weekend include:

* Bonobo loading the entire (read 80 kb) locale.alias file into 100 kb of ram for every program that uses it

* gnome-vfs initing the mime system in gnome-settings-daemon, causing ~ 160 kb of allocations, just to find out who handles `text/plain'

* Nautilus keeping two copies of the desktop background in ram (on NLD it costs about 13 mb to have a desktop background, over a plain color).

* Nautilus keeping a few 100 kb of data stored in xml files as is parsed by libxml2 (aka, bloated storage format [TM])

* The clock, wnck, and notification applet being run out of process, costing 1 MB each due to startup costs of misc libraries

The list goes on.

Lets fix this shit once 2.10 gets released.

This is great! My deepest congrats Ben!


200409   200412   200501   200502   200503   200504   200505   200506   200507   200508   200509   200510   200511   200512   200601   200602   200603   200604   200605   200606   200607   200608   200609   200610   200611   200612   200701   200702   200703   200704   200705   200707   200708   200709   200710   200711   200712   200801   200802   200803   200804   200805   200806   200807   200808   200809   200810   200811   200812   200901   200902   200903   200904   200905   200906   200907   200908   200909   200912   201001   201002   201003   201004   201007   201009   201011   201102  

This page is powered by Blogger. Isn't yours?

Subscribe to Posts [Atom]